A reservoir of time constants for memory traces in cortical neurons

Alberto Bernacchia, Hyojung Seo, Daeyeol Lee, Xiao-Jing Wang

Research output: Contribution to journalArticle

Abstract

According to reinforcement learning theory of decision making, reward expectation is computed by integrating past rewards with a fixed timescale. In contrast, we found that a wide range of time constants is available across cortical neurons recorded from monkeys performing a competitive game task. By recognizing that reward modulates neural activity multiplicatively, we found that one or two time constants of reward memory can be extracted for each neuron in prefrontal, cingulate and parietal cortex. These timescales ranged from hundreds of milliseconds to tens of seconds, according to a power law distribution, which is consistent across areas and reproduced by a 'reservoir' neural network model. These neuronal memory timescales were weakly, but significantly, correlated with those of monkey's decisions. Our findings suggest a flexible memory system in which neural subpopulations with distinct sets of long or short memory timescales may be selectively deployed according to the task demands.

Original languageEnglish (US)
Pages (from-to)366-372
Number of pages7
JournalNature Neuroscience
Volume14
Issue number3
DOIs
StatePublished - Mar 2011

Fingerprint

Reward
Neurons
Haplorhini
Parietal Lobe
Neural Networks (Computer)
Gyrus Cinguli
Prefrontal Cortex
Decision Making
Learning

ASJC Scopus subject areas

  • Neuroscience(all)

Cite this

A reservoir of time constants for memory traces in cortical neurons. / Bernacchia, Alberto; Seo, Hyojung; Lee, Daeyeol; Wang, Xiao-Jing.

In: Nature Neuroscience, Vol. 14, No. 3, 03.2011, p. 366-372.

Research output: Contribution to journalArticle

Bernacchia, Alberto ; Seo, Hyojung ; Lee, Daeyeol ; Wang, Xiao-Jing. / A reservoir of time constants for memory traces in cortical neurons. In: Nature Neuroscience. 2011 ; Vol. 14, No. 3. pp. 366-372.
@article{46e14af26f2b4f1eb728c7a914753827,
title = "A reservoir of time constants for memory traces in cortical neurons",
abstract = "According to reinforcement learning theory of decision making, reward expectation is computed by integrating past rewards with a fixed timescale. In contrast, we found that a wide range of time constants is available across cortical neurons recorded from monkeys performing a competitive game task. By recognizing that reward modulates neural activity multiplicatively, we found that one or two time constants of reward memory can be extracted for each neuron in prefrontal, cingulate and parietal cortex. These timescales ranged from hundreds of milliseconds to tens of seconds, according to a power law distribution, which is consistent across areas and reproduced by a 'reservoir' neural network model. These neuronal memory timescales were weakly, but significantly, correlated with those of monkey's decisions. Our findings suggest a flexible memory system in which neural subpopulations with distinct sets of long or short memory timescales may be selectively deployed according to the task demands.",
author = "Alberto Bernacchia and Hyojung Seo and Daeyeol Lee and Xiao-Jing Wang",
year = "2011",
month = "3",
doi = "10.1038/nn.2752",
language = "English (US)",
volume = "14",
pages = "366--372",
journal = "Nature Neuroscience",
issn = "1097-6256",
publisher = "Nature Publishing Group",
number = "3",

}

TY - JOUR

T1 - A reservoir of time constants for memory traces in cortical neurons

AU - Bernacchia, Alberto

AU - Seo, Hyojung

AU - Lee, Daeyeol

AU - Wang, Xiao-Jing

PY - 2011/3

Y1 - 2011/3

N2 - According to reinforcement learning theory of decision making, reward expectation is computed by integrating past rewards with a fixed timescale. In contrast, we found that a wide range of time constants is available across cortical neurons recorded from monkeys performing a competitive game task. By recognizing that reward modulates neural activity multiplicatively, we found that one or two time constants of reward memory can be extracted for each neuron in prefrontal, cingulate and parietal cortex. These timescales ranged from hundreds of milliseconds to tens of seconds, according to a power law distribution, which is consistent across areas and reproduced by a 'reservoir' neural network model. These neuronal memory timescales were weakly, but significantly, correlated with those of monkey's decisions. Our findings suggest a flexible memory system in which neural subpopulations with distinct sets of long or short memory timescales may be selectively deployed according to the task demands.

AB - According to reinforcement learning theory of decision making, reward expectation is computed by integrating past rewards with a fixed timescale. In contrast, we found that a wide range of time constants is available across cortical neurons recorded from monkeys performing a competitive game task. By recognizing that reward modulates neural activity multiplicatively, we found that one or two time constants of reward memory can be extracted for each neuron in prefrontal, cingulate and parietal cortex. These timescales ranged from hundreds of milliseconds to tens of seconds, according to a power law distribution, which is consistent across areas and reproduced by a 'reservoir' neural network model. These neuronal memory timescales were weakly, but significantly, correlated with those of monkey's decisions. Our findings suggest a flexible memory system in which neural subpopulations with distinct sets of long or short memory timescales may be selectively deployed according to the task demands.

UR - http://www.scopus.com/inward/record.url?scp=79952006391&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79952006391&partnerID=8YFLogxK

U2 - 10.1038/nn.2752

DO - 10.1038/nn.2752

M3 - Article

VL - 14

SP - 366

EP - 372

JO - Nature Neuroscience

JF - Nature Neuroscience

SN - 1097-6256

IS - 3

ER -